Use of Winemaking Supplements To Modify the Composition and

Feb 1, 2017 - ... Council Training Centre for Innovative Wine Production and ‡School of Agriculture, .... were purchased either from Sigma-Aldrich (...
0 downloads 0 Views 1MB Size
Subscriber access provided by Fudan University

Article

Use of Winemaking Supplements to Modify the Composition and Sensory Properties of Shiraz Wine Sijing Li, Keren A. Bindon, Susan Elaine Putnam Bastian, Vladimir Jiranek, and Kerry L. Wilkinson J. Agric. Food Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jafc.6b04505 • Publication Date (Web): 01 Feb 2017 Downloaded from http://pubs.acs.org on February 5, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of Agricultural and Food Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 39

Journal of Agricultural and Food Chemistry

Use of Winemaking Supplements to Modify the Composition and Sensory Properties of Shiraz Wine

Sijing Li1,2, Keren Bindon3, Susan E. P. Bastian2, Vladimir Jiranek1,2, and Kerry L. Wilkinson1,2*

1

The Australian Research Council Training Centre for Innovative Wine Production.

2

The University of Adelaide, School of Agriculture, Food and Wine, PMB 1, Glen Osmond, SA

5064, Australia 3

The Australian Wine Research Institute, PO Box 197, Glen Osmond, SA 5064, Australia

*

Corresponding Author: Dr Kerry Wilkinson, telephone: + 61 8 8313 7360 facsimile: + 61 8 8313

7716, email: [email protected]

1

ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 2 of 39

1

Abstract

2

Wine quality can be significantly impacted by tannin and polysaccharide composition, which can in

3

turn be influenced by grape maturity and winemaking practices. This study explored the impact of

4

three commercial wine additives, a maceration enzyme, an oenotannin and a mannoprotein, on the

5

composition and sensory properties of red wine; in particular, in mimicking the mouthfeel

6

associated with wines made from riper grapes. Shiraz grapes were harvested at 24 °Brix and 28

7

°Brix and the former vinified with commercial additives introduced either individually or in

8

combination. Compositional analyses of finished wines included tannin and polysaccharide

9

concentration, composition and size distribution by high performance liquid chromatography, while

10

the sensory profiles of wines were assessed by descriptive analysis. As expected, wines made from

11

riper grapes were naturally higher in tannin and mannoprotein than wines made from grapes

12

harvested earlier. Enzyme addition resulted in a significantly higher concentration and average

13

molecular mass of wine tannin, which increased wine astringency. Conversely, mannoprotein

14

addition reduced tannin concentration and astringency. Addition of oenotannin did not meaningfully

15

influence wine composition or sensory properties.

16 17

Key words: maceration enzyme, mannoprotein, mouthfeel, polysaccharide, sensory evaluation,

18

tannin, wine

19

2

ACS Paragon Plus Environment

Page 3 of 39

Journal of Agricultural and Food Chemistry

20

INTRODUCTION

21

Tannin and polysaccharide are the two most abundant macromolecules in red wine. Red wine

22

tannin is predominantly condensed tannin (i.e. condensation polymers of flavan-3-ols), which are

23

extracted from grape skin and seeds during alcoholic fermentation. Tannins are diverse in terms of

24

both size, typically expressed as mean degree of polymerisation (mDP), and subunit composition.1

25

Tannin is the major contributor to astringent mouthfeel in wine, which is considered central to wine

26

sensory ‘structure’,2 and this sensory effect depends on concentration, mDP and subunit

27

composition.3,4 As such, tannin strongly influences wine quality, evidenced by a recent survey of

28

seventy-three Australian Shiraz wines, which found a positive relationship between wine quality

29

grades and tannin concentration, higher mDP and the molar fraction of epigallocatechin as an

30

extension subunit.5

31

The polysaccharides typically found in wine include: those extracted from the pectic

32

network of grape cell walls, i.e. rhamnogalacturonan (RG) II and polysaccharides rich in arabinose

33

and galactose (PRAGs), such as arabinogalactan-proteins and arabinans; as well as mannoproteins

34

(MP), which result from the decomposition of yeast cells during fermentation and aging on lees.6

35

Polysaccharides interact with polyphenolic compounds and therefore play an important role in

36

modifying the taste and mouthfeel properties of wine.2 The major wine polysaccharides (RGs,

37

PRAGs and MPs) have been shown to be negatively associated with bitterness and astringency in

38

red wine.4,7 Polysaccharides may also directly contribute to wine mouthfeel properties, although this

39

effect has not been extensively studied in red wine. Purified polysaccharides (RGs, PRAGs and

40

MPs) have been found to significantly enhance the perception of palate fullness, when added to

41

model wine, however this effect was absent when polysaccharides were combined with grape

42

tannins at levels typical for red wine.2,8 In this manner, tannins and polysaccharides, individually

43

and through their interactions, may play important roles in determining the sensory (textural)

44

attributes of red wine.

3

ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 4 of 39

45

Fruit maturity at harvest has a significant impact on the macromolecule composition of

46

wine. Bindon and colleagues7,9 showed that wines made from Cabernet Sauvignon grapes harvested

47

earlier, i.e. at lower total soluble solids (TSS) content, had lower tannin and MP concentrations, and

48

less apparent astringency and viscosity, than wines made from grapes harvested later. While the

49

textural attributes associated with wines made from riper grapes might be desirable from a quality

50

standpoint, the implications of taxation imposed at higher alcohol volumes for wine exports could

51

make this unprofitable. To this end, winemaking practices that can affect the concentration and

52

composition of tannins and polysaccharides and in turn mouthfeel, in particular different must

53

processing techniques, duration of maceration and choice of yeast strain, are of interest.10-12

54

However, wine tannin profile can also be modified through the use of winemaking supplements.

55

Oenotannins are commercial tannin extracts, usually prepared from grape materials or oak wood,

56

and are often added during red wine production with the intention of improving wine color and

57

mouthfeel properties.13 Increased tannin concentration can also be achieved through the application

58

of commercial pectolytic (maceration) enzymes14 although the impact on tannin subunit

59

composition and mDP has been reported to vary.15-19 A further impact of maceration enzyme use is

60

that the degradation of cell-wall derived pectin has been found to influence the polysaccharide

61

profiles of red wines, increasing the contribution of PRAGs and RG II.6,11,20 Another approach to

62

influence the polysaccharide profile of wine is through addition of yeast-derived MPs, which are

63

approved commercial wine additives. Previous studies have suggested MP addition to red wine can

64

influence polyphenol composition and improve color stabilization.21,22 However, it is interesting to

65

note that neither the impact of maceration enzymes nor MP additions on red wine sensory attributes

66

has been extensively studied.

67

The aim of this study was therefore to investigate the impact of three commercial

68

winemaking supplements, a maceration enzyme, an oenotannin and an MP, used either individually

69

or in combination, to modify the composition and sensory properties of Shiraz wine. The study also

4

ACS Paragon Plus Environment

Page 5 of 39

Journal of Agricultural and Food Chemistry

70

aimed to evaluate whether or not these supplements might be used to mimic the desirable mouthfeel

71

properties typically associated with riper fruit, in wines made from earlier harvested grapes.

72 73

5

ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 6 of 39

74

MATERIALS AND METHODS

75

Chemicals. Reagents and reference compounds (≥ 97% purity) used for GC-MS and HPLC

76

analyses were purchased either from Sigma Aldrich (NSW, Australia) or Alfa Aesar (MA, USA).

77

Deuterium-labelled internal standards were sourced from CDN Isotopes (Pointe-Claire, QC,

78

Canada) and analytical grade sodium chloride and HPLC solvents from Chem-Supply (Gillman,

79

SA, Australia) and Merck (Kilsyth, Victoria, Australia), respectively.

80 81

Winemaking Trials. Shiraz grapes were sourced from a commercial vineyard located in the

82

McLaren Vale region of South Australia (35°17’S, 138°55’E). The regional climate is considered to

83

be temperate-warm; the average mean January temperature (MJT) for the site was 22.9 °C, based on

84

climate data from 2000 to 2016, obtained for Noarlunga (35°16’S, 138°51’E) from the Australian

85

Bureau of Meteorology (www.bom.gov.au). In 2015 the MJT was 22.1°C, with mean maximum

86

temperatures exceeding 30 °C on 8 days (data not shown). Grapes were harvested at two distinct

87

time points: (i) Harvest A (5th Feb 2015), when the total soluble solids (TSS) and pH of grapes (270

88

kg) were 23.9 °Brix and 3.4 respectively; and (ii) Harvest B (11th Feb 2015), when the TSS and pH

89

of grapes (40 kg) were 27.7 °Brix and 3.7 respectively. Berry weights were measured (200 berries,

90

in triplicate) at each time point, and were found to be significantly different (P = 0.024), being 0.96

91

and 0.90 for Harvest A and Harvest B respectively, suggesting a small amount of berry shrivel may

92

have occurred. Following harvest, grapes were chilled overnight at 0 °C, before being distributed

93

into parcels (comprising 12 kg each). Grape parcels were destemmed and crushed, with the addition

94

of 80 mg/L PMS (potassium metabisulfite, 10% solution, where the addition rate was based on 50%

95

juice yield assumption). Harvest A grape parcels were randomly allocated (in triplicate) to seven

96

different winemaking treatments (Figure 1). Control wines were made by inoculating must with 30

97

g/hL EC1118 yeast (Lallemand, SA, Australia) immediately after crushing and fermented at 28 °C,

98

with the cap plunged twice daily. Diammonium phosphate (DAP) additions (200 mg/L) were made

99

on the first and third days of fermentation. After 7 days, wines were pressed, transferred into 5 L 6

ACS Paragon Plus Environment

Page 7 of 39

Journal of Agricultural and Food Chemistry

100

glass demijohns and fermented to dryness, i.e. to < 1 g/L residual sugar, determined enzymatically

101

(D-fructose/D-glucose assay kit, Megazyme, USA). Wines were racked from gross lees and 60

102

mg/L of PMS added prior to cold stabilization (at 0 °C for three months). Wines made with the

103

addition of mannoprotein (hereafter ‘MP’) were prepared as per the control, but with the addition of

104

300 mg/L of a commercial mannoprotein after racking from gross lees. Cold soak treatment

105

involved must being stored at 10 °C for 61 hours prior to inoculation and fermentation as per the

106

control, and included: wines made without any winemaking supplements (hereafter ‘cold soak’);

107

wines made with the addition of a commercial maceration enzyme (72.6 mg/kg fruit) prior to

108

inoculation (hereafter ‘enzyme’); and wines made with the addition of both the maceration enzyme

109

and the mannoprotein (as above, hereafter ‘enzyme + MP’). The cold soak wines serve as an

110

additional control group to wines made with enzymes, accounting for any effects and variations

111

potentially arising from the cold soak process. Wines made with the addition of tannin (hereafter

112

‘tannin’), involved must being inoculated and fermented as per the control, but with the addition of

113

400 mg/L of a commercial oenotannin during the first punch down (according to manufacturer

114

instructions). Wines were also made with the addition of both tannin and mannoprotein (hereafter

115

‘tannin + MP’). Harvest B grape parcels (in triplicate) were inoculated and fermented as per the

116

control, to produce wines expected to contain naturally higher levels of tannin and mannoprotein, as

117

well as higher alcohol by volume (abv) content (hereafter ‘later harvest’). Following cold

118

stabilization, all wines were again racked from lees and 30 mg/L of PMS added; no wines

119

underwent malolactic fermentation. Wines derived from Harvest A fruit did not require pH

120

adjustment, but the pH of Later Harvest wines was adjusted to 3.6 by addition of 2 g/L tartaric acid.

121

Wines were then bottled (under screw cap closures) and stored at 15 °C for three months, prior to

122

chemical and sensory analysis. The commercial additives used in this study were sourced from

123

Laffort Australia. Additive composition (as reported by the manufacturer) follows: the enzyme

124

preparation, Lafase HE Grand Cru, contained predominately polygalacturonase activity with

125

arabinose side activity; the oenotannin, a 1:1 blend of VR COLOR and VR SUPRA, were extracted 7

ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 8 of 39

126

solely from grape material; and the mannoprotein, Oenolees MP, comprised a yeast cell wall

127

extract.

128 129

Basic Wine Composition. Wine ethanol concentrations were determined using an alcolyzer (Anton

130

Paar, Graz, Austria). The pH and titratable acidity (TA, as g/L equivalents of tartaric acid) of wines

131

were measured by the Australian Wine Research Institute’s (AWRI) Commercial Services

132

Laboratory using a TitraMaster (Hach, Victoria, Australia). Wine color density, total phenolics and

133

anthocyanins were determined by modified Somers Color assay.23 The glucose, fructose, glycerol

134

and malic acid content of wines were measured by HPLC (Agilent 1100 HPLC, Agilent

135

Technologies Australia Pty. Ltd., Melbourne, Australia) fitted with an Aminex HPX-87H column

136

(300 mm x 7.8 mm, BioRad, California, USA). Wines were filtered through 0.45 µm PVDF syringe

137

filters, (Merck Millipore, Co. Cork, IRL) before being injected into the HPLC. The mobile phase

138

was 2.5 mM H2SO4, with a flow rate of 0.5 mL/min for a 35 min run time, at 60 ºC. Signals were

139

detected with an Agilent G1315B Diode Array detector and an Agilent G1362A reflective index

140

detector. Instrument control and data analysis were performed with ChemStation software (version

141

B.01.03). Quantitation was achieved using external calibration curves (R2 > 0.99) prepared from

142

authentic standards.

143 144

Tannin Analysis. Total wine tannin concentrations were determined using the MCPT assay,23 and

145

tannins were also isolated by solid phase extraction (SPE)24 with previously described

146

modifications,5 reconstituted in methanol (to a final concentration of 10 g/L) and analyzed by

147

phloroglucinolysis25 to determine subunit composition, mDP and molecular mass, with

148

modifications described previouly.26 For both the MCPT assay and phloroglucinolysis, (-)-

149

epicatechin (Sigma Aldrich, St. Louis, MO, USA) was used as a standard for quantitation, as

150

described elsewhere.23,25 Tannin size distribution was determined by gel permeation

151

chromatography (GPC),26 with an Agilent 1200 HPLC. Tannin fractions isolated by SPE were 8

ACS Paragon Plus Environment

Page 9 of 39

Journal of Agricultural and Food Chemistry

152

diluted 1:5 with the HPLC mobile phase prior to injection. The instrumentation, chromatographic

153

conditions and calibrations for GPC analysis were the same as those described previously.27

154 155

Polysaccharide Analysis. Wine soluble polysaccharides were precipitated by diluting wine (at a

156

1:5 ratio) with ethanol and refrigerating overnight at 4 °C. Precipitates were recovered by

157

centrifugation (3273g for 5 min) and dialyzed against 4 changes of Milli-Q water using a Pur-A-

158

Lyzer Midi 3500 Dialysis kit (Sigma Aldrich, St. Louis, MO, USA), before being freeze-dried. The

159

resulting polysaccharide extracts were dissolved in Milli-Q water and hydrolyzed in 2 M

160

trifluoroacetic acid for 3 h at 100 °C. Hydrolysates were dried in vacuo and reconstituted in 0.4 mL

161

Milli-Q water and mixed 1:1 with an aqueous internal standard solution comprising 0.6 mM ribose

162

and deoxy-glucose (Sigma Aldrich, St. Louis, MO, USA). Mixtures were derivatized with 1-

163

phenyl-3-methyl-5-pyrazolone (PMP) and analyzed by HPLC using the method described

164

previouly.28 For determination of free monosaccharides, 0.06 g of polyvinylpolypyrrolidone (PVPP)

165

was added to a 1 mL aliquot of wine and the resulting mixture vortexed and centrifuged (75000g for

166

5 min). The supernatant obtained was then mixed 1:1 with the internal standard and analyzed, as

167

above.

168 169

Sensory Analysis. Descriptive analysis (DA) was undertaken approximately three months after

170

bottling. The panel comprised four male and seven female judges, with all but one panelist being

171

recruited from a pool of wine science researchers from the University of Adelaide; the majority of

172

whom had extensive DA experience. The DA process followed consensus-based methodology.29

173

Panelists attended five two-hour training sessions during which they generated appropriate

174

descriptive terms, their definitions and reference standards, based on a variety of Shiraz wines,

175

including wines from the current study. During the first two training sessions, panelists evaluated

176

wine appearance, aroma and palate characteristics and as a group discussed the scale for each

177

attribute. Wine appearance terms (color and depth) were subsequently removed since the whole 9

ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 10 of 39

178

panel agreed that they did not perceive any differences between wines after at least one replicate

179

from each treatments were presented. In subsequent training sessions, panelists were presented with

180

standards for aroma attributes, for astringency, hotness and palate fullness, and for palate

181

coarseness (‘touch standards’ comprising of a range of fabrics and powders). After gaining

182

familiarity with the standards, panelists rated wine sensory properties as a group and compared

183

results. A practice session was then held in sensory booths to evaluate judge performance, using

184

PanelCheck software (v1.4.0). In the subsequent session, feedback was provided to panelists and

185

some non-discriminating terms were discussed and removed based on panel consensus. The panel

186

eventually decided on an attribute list comprising 9 aroma, 8 flavor, 4 taste and 4 mouthfeel terms

187

(Supplementary Table 1). During formal evaluation, panelists rated the intensity of these attributes

188

on 15 cm unstructured line scales, with anchors at 10%, 50% and 90% of the scale corresponding to

189

‘low’, ‘medium’ and ‘high’, respectively; with the exception of palate coarseness, which was

190

anchored with the terms ‘talc’, ‘chalk’ and ‘pumice’, from low to high.

191

Two formal evaluation sessions were held, during which panelists were presented with twelve wine

192

samples (50 mL) in ISO standard clear wine glasses, covered with plastic lids. Formal evaluations

193

were held in isolated booths in a dedicated sensory laboratory, maintained at 21 ºC. 24 Wines, i.e. 3

194

replicates of each treatment, were presented once to each panelist, in a randomized, balanced order.

195

Data was acquired using FIZZ software (Version 2.50a, Biosystems, Couternon, France). To avoid

196

sensory fatigue, panelists were required to rest for 1 min after each sample and for 5 min after every

197

4 samples. Panelists were provided filtered water, 1 g/L pectin solution (pectin from citrus peel,

198

Sigma-Aldrich, NSW, Australia) and plain water crackers to cleanse their palates between samples.

199 200

Data Analysis. Chemical data were subjected to analysis of variance (ANOVA) using XLSTAT

201

(Version 2015.4.1, VSN International Limited, Herts, UK). Mean comparisons were performed by

202

least significant difference (LSD) multiple comparison test at P < 0.05. PanelCheck (V1.4.2,

203

Nofima Mat) was used to evaluate panel performance during DA; principal component analysis 10

ACS Paragon Plus Environment

Page 11 of 39

Journal of Agricultural and Food Chemistry

204

(PCA) of sensory data was performed using SENPAQ (Version 6.03, Qi Statistics, Reading, United

205

Kingdom).

206

RESULTS AND DISCUSSION

207

This study explored to what extent winemaking supplements could modify wine composition,

208

especially tannin and polysaccharide concentration, and therefore the organoleptic characteristics of

209

Shiraz wine. A later harvest treatment was deliberately included to generate wines naturally higher

210

in tannin and polysaccharide (MP and PRAG),9 to enable the potential for additives to mimic

211

mouthfeel characters of wine made from riper grapes to be considered.

212 213

Impact of Winemaking Supplements on Basic Wine Composition. Significant differences in

214

basic wine composition were not observed amongst Harvest A wines, but as expected, differences

215

were apparent between Harvest A and Harvest B wines (Table 1). Notably, ethanol concentrations

216

increased from 14% for Harvest A wines, to 16.7% for Harvest B wines. Wine pH, TA and malic

217

acid concentrations were similar across all treatments of Harvest A, with only MP treatment having

218

slightly higher pH and the enzyme-treated wines having a marginally higher TA. Wines made from

219

the later harvest had significantly higher TA than Harvest A wines, which could be attributed to the

220

addition of tartaric acid at bottling. Harvest B wines also had higher levels of residual fructose and

221

glycerol, which was expected for wines made from riper grapes.9

222 223

Impact of Winemaking Supplements on Wine Polysaccharide Composition. At the time of

224

analysis (3 months after bottling) the concentrations of wine soluble polysaccharides ranged from

225

514 mg/L (for enzyme treatment) to 680 mg/L (for MP treatment), but were not significantly

226

different (p = 0.057) between treatments (Table 2). However, in the analysis conducted at the

227

completion of fermentation, similar polysaccharide concentrations were observed in all wines,

228

being 506 to 716 mg/L, except for wines derived from enzyme treatment that had significantly

229

lower total polysaccharide levels, at 384 mg/L (data not shown). The small increase in 11

ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 12 of 39

230

polysaccharide levels observed in enzyme treatments from fermentation to 3 months after bottling

231

was due to mannoprotein release by yeast during the later stages of fermentation, which will be

232

discussed below.

233

Despite only minor differences in concentration, polysaccharide composition varied

234

considerably between treatments, as evidenced by monosaccharide residues recovered following

235

acid hydrolysis of polysaccharides (Figure 2).The main monosaccharide residues found were

236

mannose, galacturonic acid, galactose and arabinose, with minor contribution of rhamnose and

237

glucose, consistent with other studies.9,11,30 Comparing the control wines with those made from the

238

later harvest grapes, the latter treatment gave higher concentrations of mannose, galactose and

239

arabinose (Figure 2). Mannose can be attributed to MPs, which are released by yeast during

240

fermentation and aging. One study9 found the MP concentration increased in wines made from

241

sequentially-harvested grapes, which the authors attributed to increased yeast turnover in the higher

242

sugar musts. Galactose and arabinose residues are usually attributed to polysaccharides rich in

243

arabinose and galactose (PRAGs), which are either readily soluble, or cleaved from the pectic cell

244

wall fraction. It is generally accepted that during ripening, the PRAG and homogalacturonan (HG)

245

fractions of cell walls are depolymerized and solubilized, although this tends to occur during the

246

early stages of grape development31 and does not necessarily translate into increased grape-derived

247

polysaccharide in wines made from riper grapes.9 However, in Shiraz grapes, water-soluble

248

polysaccharides in skin cell walls have been found to increase steadily over a 10 week period post-

249

veraison.31 In the current study therefore, the increasing concentration of these polysaccharide

250

residues observed in wines made from the later harvest may indicate a cultivar-dependent

251

phenomenon (Table 2).

252

As was expected, the addition of both enzyme and MP supplements noticeably altered wine

253

polysaccharide composition, while tannin additions had no significant impact (Table 2).

254

Commercial enzyme preparations usually contain pectin-degrading activity, but some also comprise

255

hemicellulase and potentially glucosidase enzymes,15,32 hence their application during red wine 12

ACS Paragon Plus Environment

Page 13 of 39

Journal of Agricultural and Food Chemistry

256

fermentation typically results in substantial degradation of the grape cell wall.14,32 The enzyme

257

supplement employed in this study primarily exhibited polygalacturonase activity, and to a lesser

258

extent, arabinase activity, according to the manufacturer; which was supported by preliminary

259

analysis of its effect on polysaccharides purified from Chardonnay juice (Supplementary Figure 1).

260

Analyses of the monosaccharide composition of enzyme-treated wine polysaccharide showed

261

significant (≤ 60%) decreases in galacturonic acid and arabinose (Figure 2), compared to the control

262

wines, indicating an extensive degradation of pectic polysaccharides of HGs20,33 and PRAGs6 (or

263

possibly other arabinose-rich polymers) in the grape cell wall. Consequently, these residues were

264

found as free monosaccharides in wine (Table 2). Notably, the high concentration of free

265

galacturonic acid in enzyme-treated wines might account for the higher TA observed (compared to

266

the cold soak control), albeit a negligible effect on pH (Table 1). Concomitant with the decrease in

267

galacturonic acid and arabinose as polysaccharide residues, the enzyme-treated wines had

268

considerably higher rhamnose as a polysaccharide residue (Figure 2). Numerous studies have also

269

found an increased proportion of rhamnose in wine polysaccharide made with addition of

270

polygalacturonase enzymes, which has been attributed to the enrichment of RG II.6,11,20,30,34 RG II is

271

resistant to degradation by commercial enzyme preparations, due to its complex structure, therefore

272

it is often released into wine intact following enzyme treatment while HGs are largely degraded.35

273

This increase of RG II could account for the relatively high level of retained galacturonic acid in the

274

polysaccharides of enzyme-treated wines. Interestingly, the enzyme treatments also resulted in a

275

higher concentration of mannose as a polysaccharide residue relative to the control and cold soak

276

treatments (Figure 2). This finding contradicts previous studies

277

further investigation.

6,11,30,34

and therefore requires

278

Addition of MP supplements gave increased concentrations of polysaccharide-derived

279

mannose (Figure 2). However mannose content didn't reflect the initial 400 mg/L MP added, nor

280

were any meaningful changes in total polysaccharide concentrations observed, which was

281

unexpected. The increased mannose concentration as a polysaccharide residue ranged from 10 to 30 13

ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 14 of 39

282

mg/L, for Enzyme + MP, Tannin + MP, and MP treatments compared to the other treatments

283

(Figure 2). We considered that the effect recovery of mannose in treated wines would be foremost

284

dependent on the purity of the product. This was tested by measuring the hydrolysis of the MP

285

product to component monosaccharide residues (in triplicate), and it was found to yield 96.04 mg/g

286

product, which accounted for slightly less than 10% of the total material. To our surprise, 123.19

287

mg/g and 128.34 mg/g of galactose and arabinose residues were also detected respectively, which

288

may explain the proportional increases in galactose and arabinose residues as wine polysaccharide

289

in the MP treatments compared to the control. However, these changes were not consistently

290

observed where MP addition was combined with other additives. Nevertheless, based on the results,

291

while marketed as a mannoprotein, the MP product used in this study likely contained a

292

considerable amount of PRAGs and it was of interest to note that the polysaccharides isolated from

293

the MP treatment were compositionally similar to that of the wine made from grapes harvested

294

later.

295 296

Impact of Winemaking Supplements on Wine Tannins and Color. In comparison to the control,

297

wines made from the later harvest grapes were significantly higher in tannin and anthocyanin

298

concentration, and differed substantially in tannin composition, having significantly higher molar

299

proportions of epigallocatechin, together with a higher tannin mDP and molecular mass. On the

300

other hand, lower molar proportions of epicatechin-gallate were observed in tannin from these

301

wines (Table 3). These differences in phenolic composition were also reflected in color

302

measurements, namely higher levels of SO2-resistant pigments and higher color density. Similar

303

trends in the ripening-induced modification of wine tannin composition and color have been

304

observed in Cabernet Sauvignon9 and Cabernet Franc10 wines.

305

Within the Harvest A treatments, enzyme and enzyme + MP had significantly higher tannin

306

concentrations, e.g. 1,478 mg/L and 1,419 mg/L respectively, compared to 1,168 mg/L for control

307

wines. Similarly, anthocyanin concentration was also higher in wines made with enzyme addition, 14

ACS Paragon Plus Environment

Page 15 of 39

Journal of Agricultural and Food Chemistry

308

although SO2-resistant pigments and color density were unaffected. Enzyme treatment gave higher

309

molar proportions of epigallocatechin as a tannin subunit than for cold soak treatment but not

310

control. The reverse trend was observed for epicatechin-gallate in comparison to control and cold

311

soak treatments. In terms of tannin polymerization degree, the enzyme treatment resulted in

312

significantly higher tannin mDP and molecular mass relative to other Harvest A treatments, being

313

comparable with later harvest wines (Table 3). To further characterize the effect of enzyme on wine

314

tannin size distribution, tannin extracts were measured by gel permeation chromatography (Figure

315

3), which showed a notable shift in the elution profile of tannin isolated from enzyme-treated wines

316

towards a higher molecular mass (earlier elution time) relative to the other treatments. In

317

comparison, the late harvest wines, which have a similar mDP to enzyme wines, showed no obvious

318

change in the early-eluted fraction but had less late-elulting material. Previous studies have shown

319

maceration enzymes can enhance tannin extraction during vinification16-19,34 and their ability to

320

improve extraction from grape seeds has also been confirmed in both a winemaking trial17 and a

321

model wine-like solution.15 Nevertheless, the effect of enzyme addition on wine tannin composition

322

and mDP is often inconsistent. Ducasse and colleagues34 found that enzyme treatment increased

323

tannin mDP, but the proportions of epigallocatechin and epicatechin-gallate as tannin subunits

324

varied by vintage and the specific enzyme used. Other studies have shown the extent to which

325

enzymes affect tannin composition and mDP may also depend on the geographical origin of fruit18

326

as well as grape variety.16

327

Tannin mDP and subunit composition can indicate the origin of tannins in wine, i.e. from

328

skin or seed. Skin tannins generally have a higher mDP and are richer in epigallocatechin subunits

329

but low in epicatechin-gallate, whereas seed tannins have higher proportions of epicatechin-gallate,

330

no epigallocatechin and lower mDP, despite marginally higher hydrodynamic volumes (size by

331

GPC) than skin tannins at a set molecular mass.26 In the enzyme-treated wines, the proportions of

332

both epigallocatechin and epicatechin-gallate increased in line with mDP, suggesting extraction was

333

enhanced from both grape skins and seeds (Table 3). However, since the molar proportion of 15

ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 16 of 39

334

epicatechin-gallate was only significantly higher than in control wines but comparable to cold soak

335

wines (as discussed above), the enhanced seed extraction was likely due to extra soaking time

336

instead of the enzyme effect. The enhanced tannin release observed in wines made with enzyme

337

addition was most likely due to their ability to degrade cell walls. The (homogalacturonan-rich)

338

pectic fraction of cell walls are thought to have a higher affinity for binding tannins than

339

hemicellulose and cellulose.28 Thus, degradation of pectin may enhance tannin extraction in two

340

ways: (i) degradation of skin and seed cell walls during maceration may facilitate diffusion of

341

phenolic compounds into wine by breaking down the major barrier, i.e. grape cell walls; and/or (ii)

342

enzymes may reduce adsorption of tannin to grape solids (cell walls),36 a process by which

343

otherwise soluble tannins are adsorbed to marc or lost as lees, at the end of fermentation.

344

Tannin addition during fermentation did not result in any significant differences in tannin

345

concentration, composition, molecular mass (determined either by phloroglucinosis or GPC) or

346

wine color measurements (Table 3). It was suspected that this was due to the low purity of the

347

tannin product used in the current project. However, when the purity was tested by dissolving the

348

product in model wine and then measuring the concentration by MCPT method, the recovery was

349

70% w/w, demonstrating the loss of tannin addition was not due to product composition.

350

Nevertheless other studies37,38 have also found a lack of effect by tannin addition during

351

fermentation. A recent review13 on oenological tannins highlighted variable implications of their

352

addition on wine color12 and it is therefore difficult to conclude whether or not tannin addition is a

353

valid approach for improving color stability since it may depend both upon the product itself and

354

the wine matrix. Factors such as oenotannin type, dosage, grape quality and vintage confound

355

research findings. Furthermore, when oenotannin is added during fermentation, they can bind to cell

356

wall material present in wine, leading to a loss of tannin (due to precipitation) shortly after addition.

357

Bautista-Ortín and colleagues studied the binding interactions of six commercial tannins with

358

purified grape cell wall material in model wine and found between 13 and 61% of tannins were

359

removed after 90 min, with higher molecular mass fractions of tannin removed preferentially.39 16

ACS Paragon Plus Environment

Page 17 of 39

Journal of Agricultural and Food Chemistry

360

In the current study, the addition of MPs in combination with enzyme or tannin supplements

361

did not significantly alter phenolic composition, compared to the addition of enzyme or tannin

362

supplements alone. Wines with MP addition alone had the lowest tannin concentration of all the

363

treatments, being 949 mg/L, but the molar proportions of epigallocatechin and epicatechin-gallate,

364

mDP and molecular mass were not different to that of control wines, enabling the suggestion that

365

some tannin was removed. MP addition significantly reduced both anthocyanins and SO2-resistant

366

pigments, and as a consequence, color density was significantly lower compared to the control

367

(Table 3). It has been shown that yeast membrane MPs are able to precipitate tannins from red wine

368

in vitro,40 which may explain the decreased tannin concentrations in the MP treated wines. However

369

reports on effects of MPs (either commercial preparations or isolates derived from different yeast

370

strains) on wine color and phenolics are inconsistent.12,21,22 The impacts of addition are seemingly

371

influenced by MP characteristics, for example the percentage of protein and size distribution, but

372

no definitive relationships have been established and further research is warranted.

373 374

Impact of Winemaking Supplements on Sensory Profiles of Wines.

375

Significant differences (p < 0.05) were found in intensity ratings of ten sensory attributes, i.e. red

376

fruit, dark fruit and confectionary aromas, jammy flavor and flavor intensity, and five palate

377

attributes: sweetness, palate fullness, astringency, surface coarseness and hotness (Table 4).

378

Principal component analysis (PCA) was performed on sensory data for these attributes and the first

379

two components accounted for 81.3% of total variance (Figure 4A). Wines were separated along the

380

first principal component (PC) according to positive associations with dark fruit aroma, flavor

381

intensity, jammy flavor, sweetness, palate fullness and hotness. The later harvest wine was most

382

strongly associated with these attributes and therefore positioned furthest to the right of the biplot.

383

Red fruit and confectionary aromas were positively correlated with one another (R2 = 0.76), but

384

were negatively correlated with dark fruit aroma (R2 = -0.84 and -0.93 respectively).Wine made

385

with tannin addition was most strongly associated with these characters. Wines with MP addition 17

ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 18 of 39

386

also showed significantly higher ratings in red fruit compared to control. Although the origin of

387

these changes were not explored in the current study, it has been shown that addition of tannin and

388

mannoprotein products have been observed to interact with wine volatiles and affect wine aroma

389

and flavor22,41. A range of volatile compounds known to be important to Australian Shiraz wines42

390

were quantified as part of this study (Supplementary Table 2), but no significant impact on wine

391

volatiles were detected for tannin and mannoprotein treatments. The second PC separated wines

392

based on astringency and palate coarseness; attributes which were highly correlated with one

393

another. This was not unexpected, given surface smoothness is a sub-quality of overall astringency

394

and inclusion of this term was intended to further differentiate the impact of treatments on the

395

sensory quality of tannins (smooth vs. coarse). Control and cold soak wines were positioned close

396

to the center of the biplot, i.e. they were not strongly associated with any particular sensory

397

attributes. In contrast, wines made with enzyme addition exhibited enhanced astringency and

398

coarseness; the addition of MPs to the enzyme treated wines failed to significantly modify the

399

astringent perception. However, the wines made with only MP addition were positioned on the

400

opposing side of PC2, with significantly lower ratings for astringency and surface coarseness.

401

Interestingly, tannin addition enhanced red fruit and confectionary aromas, whereas in comparison

402

tannin and MP addition gave wines with significantly higher ratings for dark fruit aroma, jammy

403

flavor, palate fullness and hotness; i.e. tannin + MP wines most closely resembled later harvest

404

wines, based on their relative proximity on the PCA biplot (Figure 4A). The modification of aroma

405

and flavor achieved by combined tannin and MP additions were not observed in other supplement

406

combination (i.e. control vs. MP, Enzyme or Enzyme + MP)

407

To elucidate the relationships between the chemical composition and the sensory profile of

408

the wines, some chemical measurements were incorporated as supplementary data into the PCA

409

model built with sensory data (Figure 4B) and the correlation matrices are provided as supporting

410

data (Supplementary Tables 3 and 4). Where measurements were insignificant or highly correlated,

411

e.g. tannin size (mDP and molecular mass by phloroglucinosis) or free arabinose and galactose, 18

ACS Paragon Plus Environment

Page 19 of 39

Journal of Agricultural and Food Chemistry

412

some data were excluded from PCA. Only the major monosaccharide residues from polysaccharide

413

hydrolysis were plotted to represent main polysaccharides found in wine (as discussed above). To

414

this end, the sum of arabinose and galactose (arabinose+galactose) residues was used instead of the

415

individual monosaccharides since they were highly correlated and can both be attributed to PRAGs.

416 417

The winemaking supplements used in this study had more prominent effects on the

418

macromolecule composition of wines, and therefore palate characteristics, than on aroma and

419

flavor. The PCA model (Figure 4B) suggests free monosaccharides (galactose and arabinose, in

420

particular) and the molar proportion of epicatechin-gallate influenced perceptions of astringency

421

and palate coarseness the most. These constituents are indicative of enhanced cell wall degradation,

422

and possibly greater seed extraction (as discussed above). Interestingly, mDP and the molar

423

proportion of epicatechin-gallate were found to be positively correlated with astringency, whereas

424

the molar proportion of epigallocatechin was not, in agreement with other studies.3,4 Tannin

425

concentrations were also positively correlated with astringency, as expected. It was of particular

426

interest in this study to explore if and how the composition of wine polysaccharide affects

427

astringent mouthfeel. All polysaccharide-derived monosaccharide residues incorporated into the

428

PCA model showed correlations with astringency and coarseness, i.e., galacturonic acid and

429

arabinose+galactose were highly negatively correlated with these attributes (R2 between -0.822 and

430

-0.951) while rhamnose positively correlated with these two attributes (R2

431

respectively). However mannose as a polysaccharide residue contributed only slightly to

432

astringency (R2 = 0.623) with no apparent relationship (R2 = 0.486) to palate coarseness. These

433

results somewhat contradict other studies where all major wine polysaccharides (MP, PRAG and

434

RG II) were found to bear no apparent relationship7 or were negatively correlated4 with astringency

435

in red wines. These results highlighted the highly complex and interactive effects of the

436

supplements since they modify multiple wine components at once with resultant sensory effects.

437

For example, rhamnose was positively associated with astringency and palate coarseness, probably

=

0.885 and 0.816

19

ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 20 of 39

438

due to its abundance in the polysaccharides of wines made with enzymes; however it was more

439

likely that elevated tannin concentration was the main factor associated with this attribute, rather

440

than RG II, due to co-correlation of these factors. This is particularly supported by the observation

441

that the effect of polysaccharides on astringency is minor in the presence of tannins in model wine

442

solution and RG II in fact reduce astringency.2 Another example would be that wines made with

443

MP treatment were the least astringent, most likely due to their relatively low tannin, but in that

444

instance co-correlation with arabinose+galactose (instead of mannose), since the commercial MP

445

product used in this study contained more of these two residues than mannose.

446

Hotness and palate fullness were highly correlated (R2 = 0.9) and closely associated with

447

ethanol, glycerol, and fructose. Increased concentrations of ethanol and glycerol have been shown

448

to highly correlate with hotness and viscosity in red wine.7,43 Sweetness was also correlated with

449

hotness and palate fullness, and was most closely associated with fructose concentrations, although

450

given the low concentrations present in the current wines (Table 1) this relationship is unlikely to be

451

causal. Sweetness was also, to a lesser extent, associated with higher ethanol, glycerol, pH and

452

Galactose + Arabinose residues.

453

In the current study, the extent to which selected winemaking supplements could achieve

454

sensory properties that resembled wine made from riper fruit, i.e. wine naturally richer in tannin,

455

MPs and PRAGs with deeper color and more desirable mouthfeel sensations, was evaluated.

456

Harvesting grapes at optimal maturity is critical for achieving targeted wine style; however it is not

457

always possible to harvest grapes at optimal maturity due to unexpected vintage conditions (e.g.

458

climate) and/or the availability of resources in the vineyard and/or winery or the tax implications of

459

higher wine alcohol content. Furthermore, elevated temperature during the ripening period due to

460

climate change may produce grapes with under-developed color and flavor at typical sugar levels

461

for harvesting,44 forcing winemakers to either pick grapes at high sugar levels for desired sensory

462

characters or pick early to avoid high alcohol levels in wine at a loss of quality. Anecdotally, the

463

commercial harvest of Shiraz grapes from the vineyard involved in the current study was a week 20

ACS Paragon Plus Environment

Page 21 of 39

Journal of Agricultural and Food Chemistry

464

after Harvest B, which meant the grapes harvested for the ‘later harvest’ treatment (with a TSS of

465

28 °Brix) were still considered to be slightly under optimal ripeness from an industry perspective,

466

for that particular vintage. This study aimed to provide winemakers with more tools to control and

467

modify wine macromolecule composition in the winery when the grapes are of sub-optimal quality.

468

The current study confirmed that harvest time resulted in substantial differences in wine

469

macromolecule composition, especially in terms of polyphenolic compounds. However, although

470

the maceration enzyme treatment enhanced the tannin concentration and average molecular mass, as

471

well as polysaccharide-associated mannose (MPs) in Harvest A wines to levels comparable to wines

472

made from the later harvest, the two treatments were significantly different in their sensory

473

characteristics, i.e. the enzyme treated wines were noticeably more astringent and coarse on the

474

palate (Table 5). This observation indicates that mouthfeel is likely to be impacted by other wine

475

sensory components, such as being reduced by fruitiness and sweetness,45 and these more complex

476

interactions could not be distinguished in this study. However this result accentuates the importance

477

of the mouthfeel being in balance with other wine components.

478

Aside from the well-defined effects achieved by the maceration enzyme, this study also

479

demonstrated some interesting effects of the use of commercial MP and oenotannin products, as

480

well as latent interactions between wine polysaccharide and tannin that can be further explored. MP

481

addition itself decreased the tannin concentration and astringent mouthfeel, but when it was used in

482

combination with other additives, the effects were less clear The Enzyme + MP treatment was

483

similar to enzyme used alone, while the Tannin + MP treatment had a significant impact on aroma

484

and flavor, but not on mouthfeel compared to when tannin was used alone. Unexpectedly, tannin

485

additions did not give rise to any changes in tannin concentration and composition, nor were the

486

resultant wines rated more astringent or bitter; yet there was an apparent modification of the wine

487

sensory aroma profile. There is still much yet to be studied in the interactive effects of these three

488

categories of wine supplements, especially considering that other studies also showed that their

489

effectiveness seemed to be significantly affected by nature of the additive and the natural 21

ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 22 of 39

490

composition of the wines, both of which can vary considerably from case to case.21,34,46

491

Furthermore, in the current study a great loss of added tannin and inconsistent recoveries of the

492

mannoprotein product were observed. This may be due to both the composition of the particular

493

product used and the processes of precipitation and subsequent racking during vinification. Future

494

studies could involve a broader selection of winemaking supplements and investigate their effects in

495

finished wines to minimize some of these effects.

496 497

22

ACS Paragon Plus Environment

Page 23 of 39

Journal of Agricultural and Food Chemistry

498

FUNDING SOURCES

499

This research was conducted by the Australian Research Council Industrial Transformation

500

Training Centre for Innovative Wine Production (project number IC130100005). SL is also the

501

recipient of a Wine Australia supplementary scholarship (GWR Ph1312).

502 503

ACKNOWLEDGEMENTS

504

We thank Treasury Wine Estate and Laffort Australia for their generous supply of materials for this

505

study; we thank Tertius Van Der Westhuizen, Paul Smith, Renata Ristic and Stella Kassara for their

506

assistance; and we thank the University of Adelaide students and staff who were involved in the

507

sensory descriptive analysis panel.

508 509

SUPPORTING INFORMATION

510

Kinetics of galacturonic acid and arabinose release from a Chardonnay juice polysaccharide in the

511

presence of commercial pectolytic enzyme over a 5 day period in citrate buffer, pH 3.4; Attributes

512

used for sensory analysis of Shiraz wines; Concentration (µg/L) of selected aroma volatiles in

513

Shiraz wines made with the addition of enzyme, tannin and mannoprotein (MP) supplements, either

514

individually or in combination; Correlation matrix of sensory data; Correlation matrix of chemical

515

data. This material is available free of charge via the Internet at http://pubs.acs.org.

23

ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 24 of 39

516

REFERENCES

517

(1) Smith, P. A.; McRae, J. M.; Bindon, K. A. Impact of winemaking practices on the concentration and

518

composition of tannins in red wine. Aust. J. Grape Wine Res. 2015, 21, 601–614.

519

(2) Vidal, S.; Courcoux, P.; Francis, L.; Kwiatkowski, M.; Gawel, R.; Williams, P.; Waters, E.; Cheynier, V. Use

520

of an experimental design approach for evaluation of key wine components on mouth-feel perception.

521

Food. Qual. Prefer. 2004, 15, 209–217.

522

(3) Vidal, S.; Francis, L.; Guyot, S.; Marnet, N.; Kwiatkowski, M.; Gawel, R.; Cheynier, V.; Waters, E. J. The

523

mouth-feel properties of grape and apple proanthocyanidins in a wine-like medium. J. Sci. Food Agric. 2003,

524

83, 564–573.

525

(4) Quijada-Morín, N.; Williams, P.; Rivas-Gonzalo, J. C.; Doco, T.; Escribano-Bailón, M. T. Polyphenolic,

526

polysaccharide and oligosaccharide composition of Tempranillo red wines and their relationship with the

527

perceived astringency. Food Chem. 2014, 154, 44–51.

528

(5) Kassara, S.; Kennedy, J. A. Relationship between red wine grade and phenolics. 2. Tannin composition

529

and size. J. Agric. Food Chem. 2011, 59, 8409–8412.

530

(6) Doco, T.; Williams, P.; Cheynier, V. Effect of flash release and pectinolytic enzyme treatments on wine

531

polysaccharide composition. J. Agric. Food Chem. 2007, 55, 6643–6649.

532

(7) Bindon, K.; Holt, H.; Williamson, P. O.; Varela, C.; Herderich, M.; Francis, I. L. Relationships between

533

harvest time and wine composition in Vitis vinifera L. cv. Cabernet Sauvignon 2. Wine sensory properties

534

and consumer preference. Food Chem. 2014, 154, 90–101.

535

(8) Vidal, S.; Francis, L.; Williams, P.; Kwiatkowski, M.; Gawel, R.; Cheynier, V.; Waters, E. The mouth-feel

536

properties of polysaccharides and anthocyanins in a wine like medium. Food Chem. 2004, 85, 519–525.

537

(9) Bindon, K.; Varela, C.; Kennedy, J.; Holt, H.; Herderich, M. Relationships between harvest time and wine

538

composition in Vitis vinifera L. cv. Cabernet Sauvignon 1. Grape and wine chemistry. Food Chem. 2013, 138,

539

1696–1705.

540

(10) Cadot, Y.; Caillé, S.; Samson, A.; Barbeau, G.; Cheynier, V. Sensory representation of typicality of

541

Cabernet Franc wines related to phenolic composition: Impact of ripening stage and maceration time. Anal.

542

Chim. Acta 2012, 732, 91–99. 24

ACS Paragon Plus Environment

Page 25 of 39

Journal of Agricultural and Food Chemistry

543

(11) Apolinar-Valiente, R.; Romero-Cascales, I.; Williams, P.; Gómez-Plaza, E.; López-Roca, J. M.; Ros-García,

544

J. M.; Doco, T. Effect of winemaking techniques on polysaccharide composition of Cabernet Sauvignon,

545

Syrah and Monastrell red wines. Aust. J. Grape Wine Res. 2014, 20, 62–71.

546

(12) Escot, S.; Feuillat, M.; Dulau, L.; Charpentier, C. Release of polysaccharides by yeasts and the influence

547

of released polysaccharides on colour stability and wine astringency. Aust. J. Grape Wine Res. 2001, 7, 153–

548

159.

549

(13) Versari, A.; du Toit, W.; Parpinello, G. P. Oenological tannins: a review. Aust. J. Grape Wine Res. 2013,

550

19, 1–10.

551

(14) Zietsman, A. J. J.; Moore, J. P.; Fangel, J. U.; Willats, W. G. T.; Trygg, J.; Vivier, M. A. Following the

552

compositional changes of fresh grape skin cell walls during the fermentation process in the presence and

553

absence of maceration enzymes. J. Agric. Food Chem. 2015, 63, 2798–2810.

554

(15) Bautista-Ortín, A.; Jiménez-Pascual, E.; Busse-Valverde, N.; López-Roca, J.; Ros-García, J.; Gómez-Plaza,

555

E. Effect of wine maceration enzymes on the extraction of grape seed proanthocyanidins. Food Bioprocess

556

Technol. 2013, 6, 2207–2212.

557

(16) Busse-Valverde, N.; Gomez-Plaza, E.; Lopez-Roca, J. M.; Gil-Munoz, R.; Fernandez-Fernandez, J. I.;

558

Bautista-Ortin, A. B. Effect of different enological practices on skin and seed proanthocyanidins in three

559

varietal wines. J. Agric. Food Chem. 2010, 58, 11333–11339.

560

(17) Busse-Valverde, N.; Gómez-Plaza, E.; López-Roca, J. M.; Gil-Muñoz, R.; Bautista-Ortín, A. B. The

561

extraction of anthocyanins and proanthocyanidins from grapes to wine during fermentative maceration is

562

affected by the enological technique. J. Agric. Food Chem. 2011, 59, 5450–5455.

563

(18) Moreno-Pérez, A.; Fernández-Fernández, J. I.; Bautista-Ortín, A. B.; Gómez-Plaza, E.; Martínez-Cutillas,

564

A.; Gil-Muñoz, R. Influence of winemaking techniques on proanthocyanidin extraction in Monastrell wines

565

from four different areas. Eur. Food Res. Technol. 2013, 236, 473–481.

566

(19) Ortega-Heras, M.; Pérez-Magariño, S.; González-Sanjosé, M. L. Comparative study of the use of

567

maceration enzymes and cold pre-fermentative maceration on phenolic and anthocyanic composition and

568

colour of a Mencía red wine. LWT - Food Science and Technology 2012, 48, 1–8.

25

ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 26 of 39

569

(20) Ayestarán, B.; Guadalupe, Z.; León, D. Quantification of major grape polysaccharides (Tempranillo v.)

570

released by maceration enzymes during the fermentation process. Anal. Chim. Acta 2004, 513, 29–39.

571

(21) Rodrigues, A.; Ricardo-Da-Silva, J. M.; Lucas, C.; Laureano, O. Effect of commercial mannoproteins on

572

wine colour and tannins stability. Food Chem. 2012, 131, 907–914.

573

(22) Del Barrio-Galán, R.; Pérez-Magariño, S.; Ortega-Heras, M.; Guadalupe, Z.; Ayestarán, B. Polysaccharide

574

characterization of commercial dry yeast preparations and their effect on white and red wine composition.

575

LWT - Food Science and Technology 2012, 48, 215–223.

576

(23) Mercurio, M. D.; Dambergs, R. G.; Herderich, M. J.; Smith, P. A. High throughput analysis of red wine

577

and grape phenolics - Adaptation and validation of methyl cellulose precipitable tannin assay and modified

578

somers color assay to a rapid 96 well plate format. J. Agric. Food Chem. 2007, 55, 4651–4657.

579

(24) Jeffery, D. W.; Mercurio, M. D.; Herderich, M. J.; Hayasaka, Y.; Smith, P. A. Rapid isolation of red wine

580

polymeric polyphenols by solid-phase extraction. J. Agric. Food Chem. 2008, 56, 2571–2580.

581

(25) Kennedy, J. A.; Jones, G. P. Analysis of proanthocyanidin cleavage Products following acid-catalysis in

582

the presence of excess phloroglucinol. J. Agric. Food Chem. 2001, 49, 1740–1746.

583

(26) Kennedy, J. A.; Taylor, A. W. Analysis of proanthocyanidins by high-performance gel permeation

584

chromatography. J. Chromatogr. A 2003, 995, 99–107.

585

(27) Bindon, K. A.; Bacic, A.; Kennedy, J. A. Tissue-specific and developmental modifications of grape cell

586

walls influence the adsorption of proanthocyanidins. J. Agric. Food Chem. 2012, 60, 9249–9260.

587

(28) Ruiz-Garcia, Y.; Smith, P. A.; Bindon, K. A. Selective extraction of polysaccharide affects the adsorption

588

of proanthocyanidin by grape cell walls. Carbohydr. Polym. 2014, 114, 102–114.

589

(29) Lawless, H. T.; Heymann, H. Descriptive analysis. In Sensory Evaluation of Food, 2nd ed.; Springer New

590

York: 2010; pp 227–257.

591

(30) Apolinar-Valiente, R.; Williams, P.; Romero-Cascales, I.; Gómez-Plaza, E.; López-Roca, J. M.; Ros-García,

592

J. M.; Doco, T. Polysaccharide composition of Monastrell red wines from four different spanish terroirs:

593

effect of wine-making techniques. J. Agric. Food Chem. 2013, 61, 2538–2547.

26

ACS Paragon Plus Environment

Page 27 of 39

Journal of Agricultural and Food Chemistry

594

(31) Vicens, A.; Fournand, D.; Williams, P.; Sidhoum, L.; Moutounet, M.; Doco, T. Changes in polysaccharide

595

and protein composition of cell walls in grape berry skin (Cv. Shiraz) during ripening and over-ripening. J.

596

Agric. Food Chem. 2009, 57, 2955–2960.

597

(32) Romero-Cascales, I.; Ros-García, J.; López-Roca, J.; Gómez-Plaza, E. The effect of a commercial

598

pectolytic enzyme on grape skin cell wall degradation and colour evolution during the maceration process.

599

Food Chem. 2012, 130, 626–631.

600

(33) Bonnin, E.; Dolo, E.; Le Goff, A.; Thibault, J.-F. Characterisation of pectin subunits released by an

601

optimised combination of enzymes. Carbohydr. Res. 2002, 337, 1687–1696.

602

(34) Ducasse, M. A.; Canal-Llauberes, R. M.; de Lumley, M.; Williams, P.; Souquet, J. M.; Fulcrand, H.; Doco,

603

T.; Cheynier, V. Effect of macerating enzyme treatment on the polyphenol and polysaccharide composition

604

of red wines. Food Chem. 2010, 118, 369–376.

605

(35) Doco, T.; Williams, P.; Vidal, S.; Pellerin, P. Rhamnogalacturonan II, a dominant polysaccharide in juices

606

produced by enzymic liquefaction of fruits and vegetables. Carbohydr. Res. 1997, 297, 181–186.

607

(36) Castro-López, L. d. R.; Gómez-Plaza, E.; Ortega-Regules, A.; Lozada, D.; Bautista-Ortín, A. B. Role of cell

608

wall deconstructing enzymes in the proanthocyanidin–cell wall adsorption–desorption phenomena. Food

609

Chem. 2016, 196, 526-532.

610

(37) Parker, M.; Smith, P. A.; Birse, M.; Francis, I. L.; Kwiatkowski, M. J.; Lattey, K. A.; Liebich, B.; Herderich,

611

M. J. The effect of pre- and post-ferment additions of grape derived tannin on Shiraz wine sensory

612

properties and phenolic composition. Aust. J. Grape Wine Res. 2007, 13, 30–37.

613

(38) Alcalde-Eon, C.; García-Estévez, I.; Ferreras-Charro, R.; Rivas-Gonzalo, J. C.; Ferrer-Gallego, R.;

614

Escribano-Bailón, M. T. Adding oenological tannin vs. overripe grapes: Effect on the phenolic composition of

615

red wines. J. Food Compos. Anal. 2014, 34, 99–113.

616

(39) Bautista-Ortín, A. B.; Cano-Lechuga, M.; Ruiz-García, Y.; Gómez-Plaza, E. Interactions between grape

617

skin cell wall material and commercial enological tannins. Practical implications. Food Chem. 2014, 152,

618

558–565.

27

ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 28 of 39

619

(40) Rowe, J. D.; Harbertson, J. F.; Osborne, J. P.; Freitag, M.; Lim, J.; Bakalinsky, A. T. Systematic

620

Identification of Yeast Proteins Extracted into Model Wine during Aging on the Yeast Lees. J. Agric. Food

621

Chem. 2010, 58, 2337-2346.

622

(41) Mitropoulou, A.; Hatzidimitriou, E.; Paraskevopoulou, A. Aroma release of a model wine solution as

623

influenced by the presence of non-volatile components. Effect of commercial tannin extracts,

624

polysaccharides and artificial saliva. Food Res. Int. 2011, 44, 1561-1570.

625

(42) Mayr, C. M.; Geue, J. P.; Holt, H. E.; Pearson, W. P.; Jeffery, D. W.; Francis, I. L. Characterization of the

626

key aroma compounds in Shiraz wine by quantitation, aroma reconstitution, and omission studies. J. Agric.

627

Food Chem. 2014, 62, 4528–4536.

628

(43) King, E. S.; Dunn, R. L.; Heymann, H. The influence of alcohol on the sensory perception of red wines.

629

Food. Qual. Prefer. 2013, 28, 235–243.

630

(44) Webb, L.; Whetton, P.; Barlow, E. IT Clim. Res. 2008, 36, 89–98.

631

(45) Sáenz-Navajas, M.-P.; Campo, E.; Fernández-Zurbano, P.; Valentin, D.; Ferreira, V. An assessment of the

632

effects of wine volatiles on the perception of taste and astringency in wine. Food Chem. 2010, 121, 1139-

633

1149.

634

(46) Rinaldi, A.; Gambuti, A.; Moine-Ledoux, V.; Moio, L. Evaluation of the astringency of commercial

635

tannins by means of the SDS–PAGE-based method. Food Chem. 2010, 122, 951–956.

636 637

28

ACS Paragon Plus Environment

Page 29 of 39

Journal of Agricultural and Food Chemistry

638

FIGURE CAPTIONS

639

Figure 1. Flowchart of winemaking treatments.

640 641

Figure 2. Mean monosaccharide residues following acid hydrolysis of wine soluble polysaccharides

642

in Shiraz wines made with the addition of enzyme, tannin and mannoprotein (MP), either

643

individually or in combination. Different letters indicate statistical significance determined by

644

ANOVA. Fisher’s least significant difference (LSD) was calculated post ANOVA, *p < 0.05; **

645

p